Black and Wiliam 1998
4 year study by 2 leading Professors at King‟s College
Trawled world wide for effective classroom based research
Considered 700 studies, but chose only those with large
effect sizes, classroom environment, and good design.
Test before Test after
Test before Control Group Test after
The effect size is how much better the Experimental Group
was than the Control Group
The strategies they found:
had the greatest effect on the weakest learners
could yield an effect size equivalent to a two grade leap at
Their study also found that established formative
assessment practice is weak, tending to ape summative
We have weak practice in a key variable... a real
Bill or Sid?
Which builder would you choose to build your new extension?
Sid’s High Grade Grafters
Quality Control System:
Sid grades each worker’s efforts at the end of each day
informing them of this grade
He praises work of above average standard
He draws attention to errors and deficiencies
He constructively criticises work which is not of an acceptable
He moves on to the next day’s work to guarantee speedy
Bill’s Trouble Shooters
Quality Control System:
Bill asks each worker to inspect their own work and fix errors
and deficiencies as they go
He inspects work at the end of each day, praising work of an
He gets workers to put right any errors and deficiencies and
checks these corrections have been made
He constructively criticises work which is not of an
Moves on to the next day’s work to guarantee speedy
Medal and Mission feedback
1. Make the goals very clear: criteria are explained and illustrated
2. Ask the student for a self-assessment
(they will be helped if you give them criteria to
self-assess against, and exemplars.)
3. Give non-judgmental feedback:
accept the student’s present standard
avoid competition or comparison with others, instead let them
the task, and
with themselves, (i.e. with previous work)
Feedback should be
forward looking, positive, & constructive
task centred not ego centred
medal and mission
Feedback exercise (graph)
1. “Well done John, that’s brilliant! 9/10
2. “Good graph Martha. Nice and neat. All the
points are well plotted.”
3. “Better than last week. Rather thick lines
though! Keep improvement up.”
4. “Better standard. A sharper pencil improves
accuracy. Keep improvement up!”
5. “Good scales, good line, but some points missing!!
Nice and neat.”
6. “Good standard. No title. neat writing.”
7. “Good axes, points and line.
8. “Well done, you handed it in! Please finish it
now. Keep this improvement up.”
Learning Centred Feedback in Practice. Geoff Petty
Professors Ian Black and Dylan Wiliam of Kings College London reviewed many
hundreds of research studies and showed that formative assessment has more effect on
learning than any other single factor (including prior learning).*
The following summarises the advice suggested by Black and Wiliam, and then goes
further to add some concrete suggestions for implementing their ideas.
Effective formative assessment has its most positive effect on low attainers, and few teachers
adopt good practice.
Avoid grading. Grades are consistently found to demotivate low attainers. They also
fail to challenge high attainers, often making them complacent. So avoid giving a grade or
mark except where absolutely necessary. This is not easy to do on some courses.
However it is rarely necessary, and almost never desirable, to grade every piece of work.
Use self-assessment: Ask students to self-assess, providing them with self-assessment
criteria or helping them to develop their own. See examples below. Self-assessment has
been shown to double attainment if it is used very frequently. It encourages the reflective
habit of mind essential for improvement, ensures students take responsibility for their
own learning, focusses attention on criteria for success, and increases effort and
persistence. It should be followed by action planning and the action plan points should be
followed up, it then generates excellent evidence for the Key Skill „Improve Own Learning
Give learning-centred feedback “Give a medal and a mission”
Accept the student‟s present attainment however low, without blame or disapproval.
Set about improving this by giving a:
Medal for what the student can do or has done well. Effort persistence and other good
study habits can be included in the criteria.
Mission: what the student needs to do to improve. This can be an improvement to the
existing work, or a target (feed-forward task) for the next piece of work.
Focus your feedback on the following:
tasks: e.g. provide positive comments on the completion of tasks, strengths,
criterion-referenced achievement etc. If teachers set mastery tasks** this provides
opportunities to give positive feedback to the very lowest attainers.
Meeting personal targets. If students are encouraged to self-assess and to set
themselves targets for improvement, then the teacher can comment on a student‟s
progress towards these targets.
improvements effected by the student can be positively commented on.
opportunities for improvement and constructive criticism can be given.
Feedback proformas can help teachers give such feedback in practice, examples are
shown below. In each case they need to be larger than shown to allow space for
Use the „praise sandwich‟ That is: praise; constructive criticism; then praise again.
Use Mastery Learning. This is a series of easy tests on key material set every four or six
weeks, with retests for those students who do not pass. It takes time to set this up, but it
works very well. See the „Mastery Learning‟ chapter in Teaching Today.
Use formative teaching methods that „find faults, fix, and and follow up‟ see the
handout called „formative teaching methods‟ which you can download from
http://geoffpetty.moonfruit.com. These ideas were subsequent to the research review but
are clearly linked with it.
Why not give assignments, homeworks, classwork etc with a feedback proforma like
these, (only with more space). This helps the teacher give learning-focussed rather than
grading-focussed feedback. Such proformas are not new, but are under-used. Make sure
they are bigger than shown to allow space for writing.
Generic assessment criteria
Here, as an example, is a set of negotiated criteria used to assess essays. Any generic skill
could be developed in the same way, e.g. electronic circuit design, painting, lab report
Ask students for criteria first. They will come up with most of them, and will then really
„own‟ them. Follow this with a discussion on which criteria really count and why. This is
very helpful to clarify good practice and your expectations. Each criterion needs to be
discussed, explained, and justified to the class. (I do not pretend that the criteria below are
the only, or the best criteria for essay writing, you must decide your own!)
Criteria are then used repeatedly, perhaps for every essay written on the course. Students
hand in the work already self-assessed, then the teacher assesses against the same criteria.
Ideally no grade is awarded, or if it is, it is given some weeks after this informative
feedback. Black and Wiliam‟s research review shows that if you grade students pay
attention only to this, and don‟t read your feedback.
Ruth Beard in “Teaching in Higher Education” claims that such generic criteria greatly
improve importance over a course even without self-assessment. Black and Wiliam
showed that self-assessment and informative feedback were amongst the most important
things a teacher did.
The following grids should all be much bigger in practice, to allow more space for
* Black and Wiliam (1998) “Assessment and Classroom Learning” in the
journal Assessment in Education.
**See separate handout for explanation on mastery tasks
read also “Inside the black box” at:
Essay writing assessment proforma
Criteria Self-assessment Teacher assessment
Did you relate each of your
arguments to the essay
Did you give arguments both
„for‟ and „against‟ both:
The proposition in the essay
Any major points or
conclusions you made?
Did you give enough evidence,
examples, and illustrations for
each of your arguments?
Did you prioritise the arguments
for and against, and evaluate
Did you draw a justified
conclusion related directly to the
Improvements needed for this
Targets for the next essay Self assessed target:
Assessment criteria grade Teacher, peer, or self-assessment
Methods: aim to
appropriate, and as
simple or elegant as
The principles or
formulae used are
Working: aim to
make working clear;
complete; easy to
principles or formulae
used where necessary.
Care taken: aim to
check your work for
errors, and present
It’s okay if you don’t fully understand a concept first time, learning takes time.
If this work is graded, aim to beat your own record, not someone else’s
what counts is whether you understand the problem and solution, not whether you
made any silly slips
If you got something wrong that’s fine. It’s how we learn.
You will learn from mistakes if you find out how to do it without mistakes next time,
and really understand this.
Corrective work on this exercise
(Find someone with an A for …….. and ask them to show and explain their work.)
Target for your next piece of work
Freely based on ‘Using Assessment to Raise Achievement in Mathematics’ QCA Nov 2001
Opportunities for Development
Below are examples of assignment-specific assessment proformas:
Assessment Criteria Strengths and development
Plan for improving I like your ideas on diet exercise and
Health and entertainment. Most points well covered.
well-being A well designed table! Some rest would
help. Read assignment brief carefully!
Purpose for this plan You explain this well referring to
evidence. Quite the best bit of your
General Comments. I notice some of your work is neater, keep
Student‟s goals: Comments:
Be better at checking You have definitely achieved an
my spelling improvement here Simon.
Criteria Student assessment Teacher assessment
A diagram of the heart:
accurate, neat, and
Explanation of how the
valve sequence, and
Vocational Skills – Care
Self-assessment of key criteria for written work
Below is a list of some of the most important skills needed when completing any written tasks.
These skills will help you achieve your Care Modules and will also help you in the future whenever
you need to find out information for yourself and present it well.
Please think about each skill carefully and assess how well you think you do. Then score each one
Red: Needs improving
Skill Red Amber Green Teacher assessment
Using only relevant
Writing in your own
Using the library
Using computers and
Keeping a record of
sources of information
eg. books, internet
articles, leaflets, etc to
add interest to your
Handing work in on
Which of these skills do you need to improve most?
Learning Target for next piece of work:
Assessment of Writing Skills
Please think about each skill carefully and assess how well you think you do. Then score
each one as follows. Please hand in your plan with your finished work.
Skill Didn’t I think I did Teacher Assessment
Used sentences well
Used paragraphs well
Used verbs well
Used capital letters well
Used full stops and commas
Answered the question
Corrections needed for this piece of work:
Targets for next written assignment:
English Literature AS
Unit 3 Assessment: 20th Century
AO Criteria Comment
Grasps significance of
opinions in title
Uses more than one
Confident judgement of
Relevant response well-
1 focused on task
2 Methodical, structured
Develops an argument
Focuses on presentation
Mark: MEG: Target mark:
David Rowbottom firstname.lastname@example.org
Self Assessment of Language Skills
For each substantial piece of work mark the number of times each error has occurred and use this
to action plan your improvement.
Type of error Homework date and title/ Number of errors of
past part. agreement
gender of noun
use of le, la, un, une
Adjective in wrong place
Not a French word!
use of negative
order of pronouns
Pronoun in the wrong place
ue of qui, que etc
Failure to spot ce qui etc
use of subjunctive
use of imperative
Write other error types below!
Butler 1988: Able
Experimental group: 1/2 in the top quartile
1/2 in the bottom quartile
Were given tasks: (tasks were new to students)
some tasks were divergent some convergent
tasks were given over three separate sessions
grading criteria were given beforehand
Tas k 1 feedback 1 Tas k 2 feedback 2 Tas k 3 feedback 3
1/3 of students were given comments only
1/3 of students were given only grades
1/3 of students were given grades + comments.
Who learned best?
comments only: performance improved by 33%
grades only: performance declined significantly
comments + grades: performance declined, particularly on
the convergent task
Who maintained interest?
low achievers: interest was undermined by grading (with or
high achievers: interest was sustained at a high level
Moral: grading degrades learning. But students need grades! So: withhold grades and
give them alltogether every 6-12 weeks? Give students their grades four weeks ‘late’?
Discussion of Black and Wiliam‟s review
How do Black and Wiliam‟s findings differ from conventional practice?
Conventional practice: “teach, test, grade, and move on” (assessment is summative)
Too often the teacher teaches a topic, sets some work, grades it, criticises it in a more or less
constructive way, but does not check that the student has made good any deficiencies. Then the
teacher moves on to the next topic.
A common assumption behind this approach is that learning quality and quantity depend on
talent or ability, and that the role of assessment is to measure this ability. If learning is wanting,
this is attributed to a lack of ability, flair or intelligence.
Best Practice: “find faults and fix” (assessment is diagnostic)
Black and Wiliam‟s review suggests a different approach. A topic is taught, and some work is set.
The student and the teacher use this work to diagnose deficiencies and set targets for
improvement. This improvement is monitored.
The assumption is that learning quality and quantity depends on time and effort spent on
improvement, and that the role of assessment is to diagnose deficiencies so that time and effort can
be focussed on improvement. Indeed weaknesses are the very areas where greatest improvement
can be made with least effort. If learning is wanting, you need to try harder for longer.
Common Practice: teach, test, grade, and move on:
Criticise this “You can‟t fatten a pig
by weighing it”
Can do Grade that
Best Practice: “Find faults and fix”
100 % Can‟t do Self-assessment
Find this and
Medals and missions
get the student Corrections
to fix it Targets for next
Can do piece of work
Praise this Mastery learning
Research is very much in favour of the „time and effort‟ assumption rather than the talent
assumption, even in areas like music where you might expect talent to be important. „Talent‟ is
often a by-product of how much time and effort has been spent on learning in the past, due
perhaps to intense interest.
Teachers in the East, for example in Pacific Rim countries, are puzzled by the Western obsession
with talent. Their very successful education systems are built firmly on the „find faults and fix‟
Find a summary by Black and Wiliam on their research, with full references at:
Geoff Petty 2001
Judgmental Feedback Informative Feedback
“Here is my measurement” “These are your goals, this is what you
do well, and this is how to get better”
Characteristics of this Feedback
The feedback compares students with each There are clear assessment criteria and
other, and encourages them to compete. It is goals. Feedback consists of information about
„norm referenced‟. the extent to which these have been met.
The teacher gives grades, marks, and There are:
comments that make conscious or Medals: for what they have done well
unconscious comparisons with others Missions: showing how to improve
Effect on self-esteem
Judgement makes students nervous and The student feels accepted, and that their
protective of their self-esteem. So students efforts are being recognised and valued
avoid risks and challenges. Self esteem and commitment tends to rise
The self-esteem of high achieving students and there is increased emotional
involvement in tasks
Consequent learning strategies
Effective learning strategies
Maladaptive learning strategies
Deep learning. Their eyes are on the goals,
Surface learning is likely. Their eye is on the
assessment criteria, tasks, and their
grade, not understanding, learning or the
„missions‟. High quality learning aimed at
task. The student memorises, seeks short
understanding and improvement.
cuts, copies etc.
As esteem comes from effort, not
Right answer syndrome
comparative attainment, students are
prepared to take risks and accept challenges.
Student's‟ learning theory
Maladaptive and blaming learning theory Adaptive and blame free learning theory
„mistakes are shameful‟ „Effort is the key and it‟s up to me‟
„effort shows you must be stupid‟ mistakes are informative feedback
„ability is the key and it is inborn‟ Instrinsic motivation: Learning is an end in
Extrinsic motivation: It‟s only worth working if itself
you get something out of it
Effect on low achievers
There is reduced: interest, effort, There is increased:interest, effort, persistence,
persistence, self-esteem and self-belief, and self-esteem and self-belief.
less emotional investment in learning
In time: Learned resourcefulness: “There
In some cases: “Learned helplessness” must be a way round my difficulties and if I find
“No matter what I do I‟m bound to fail” it I will succeed.”
The student withdraws and retires hurt, “Learning depends on time, effort, corrected
rejecting the teachers, college, etc practice, and using the right strategies.”
Hostility towards learning Identification with the aims of the course
Learning is seen as something for others. Learning is seen as an end in itself
This document summarises some key ideas in the Black and Wiliam review 1998
“Assessment and classroom learning‟ in the journal „Assessment in Education‟